sydbarrett74 has quit [Quit: Going offline, see ya! (www.adiirc.com)]
Bogdoslav has quit [Quit: Nettalk6 - www.ntalk.de]
<tadzik>
yeah
<Stskeeps>
such a douche move to do it near thanksgiving..
<tadzik>
on one hand, as a non-american I find this quite annoying. On the other, it's comforting to now that now the entire rest of the world knows how shitty this is and will be able to recognize it
yuhl_ has quit [Quit: yuhl_]
leavehouse has quit [Ping timeout: 248 seconds]
edrex has quit [Ping timeout: 258 seconds]
Steverman has joined #ipfs
<jcgruenhage>
Even as a non-american, if the FCC pushes this through, we are all affected by this. There is a shitload of traffic going through the USA.
Steverman has quit [Ping timeout: 264 seconds]
dhruvbaldawa has joined #ipfs
dhruvbaldawa has quit [Remote host closed the connection]
dhruvbaldawa has joined #ipfs
screensaver has quit [Remote host closed the connection]
screensaver has joined #ipfs
lexruee has quit [Ping timeout: 250 seconds]
lexruee has joined #ipfs
<toppler>
I'm very confused. I'm on a company LAN whose router is apparently not configured for upnp, but ipfs is still managing to peer.
dhruvbaldawa has quit [Remote host closed the connection]
sydbarrett74 has joined #ipfs
sydbarrett74 has quit [Remote host closed the connection]
f0i has quit [Ping timeout: 240 seconds]
dhruvbaldawa has joined #ipfs
dhruvbaldawa has quit [Remote host closed the connection]
dhruvbaldawa has joined #ipfs
anewuser has joined #ipfs
jkilpatr has joined #ipfs
cwahlers_ has quit [Ping timeout: 240 seconds]
cwahlers has joined #ipfs
dhruvbaldawa has quit [Remote host closed the connection]
xnbya has quit [Ping timeout: 258 seconds]
xnbya has joined #ipfs
}ls{ has joined #ipfs
rickystef has joined #ipfs
dimitarvp has joined #ipfs
ecloud is now known as ecloud_wfh
rickystef has quit [Quit: rickystef]
cpey has joined #ipfs
clickjack has quit [Quit: clickjack]
anewuser has quit [Ping timeout: 248 seconds]
<cpey>
ipns resolution through ipfs.io seems not to be working
Jesin has quit [Quit: Leaving]
qben[m] has joined #ipfs
Steverman has joined #ipfs
dexteruk has quit [Ping timeout: 248 seconds]
<Steverman>
If a browser gets a file through js-ipfs, do I have to add it explicitly in order to let it share between other browsers?
nilbot has joined #ipfs
dexteruk has joined #ipfs
rickystef has joined #ipfs
toppler has left #ipfs ["ERC (IRC client for Emacs 25.3.1)"]
dhruvbaldawa has joined #ipfs
BioSpider has quit [Ping timeout: 240 seconds]
upperdeck has quit [Ping timeout: 240 seconds]
f0i has joined #ipfs
upperdeck has joined #ipfs
plexigras has joined #ipfs
mtodor has joined #ipfs
lupi has joined #ipfs
rickystef has quit [Quit: rickystef]
larpanet has joined #ipfs
plexigras has quit [Quit: WeeChat 1.9.1]
plexigras has joined #ipfs
anewuser has joined #ipfs
upperdeck has quit [Ping timeout: 240 seconds]
upperdeck has joined #ipfs
chiui has joined #ipfs
toppler has joined #ipfs
<toppler>
I'm still not sure what's going on with my network. I'm on a company LAN. When I start IPFS, dmesg is telling me that my firewall is dropping SYN packets destined for port 4001. I do not know how those packets got routed to my machine in the first place.
ticao[m] has joined #ipfs
Jesin has joined #ipfs
ticao[m] has left #ipfs ["User left"]
dhruvbaldawa has quit [Remote host closed the connection]
dhruvbaldawa has joined #ipfs
ashark has joined #ipfs
dhruvbaldawa has quit [Ping timeout: 248 seconds]
anewuser has quit [Quit: anewuser]
xnbya has quit [Ping timeout: 258 seconds]
xnbya has joined #ipfs
domanic has joined #ipfs
hacdias has joined #ipfs
Alpha64 has joined #ipfs
dhruvbaldawa has joined #ipfs
ashark_ has joined #ipfs
dhruvbaldawa has quit [Ping timeout: 248 seconds]
ashark has quit [Ping timeout: 240 seconds]
The_8472 has quit [Ping timeout: 250 seconds]
The_8472 has joined #ipfs
upperdeck has quit [Ping timeout: 264 seconds]
upperdeck has joined #ipfs
<Icefoz_>
toppler: MAGIC
f0i has quit [Ping timeout: 260 seconds]
SuprDewd has quit [Read error: Connection reset by peer]
SuprDewd has joined #ipfs
ulrichard has quit [Remote host closed the connection]
larpanet has quit [Ping timeout: 248 seconds]
<Icefoz_>
Okay that's not true.
<Icefoz_>
Does IPFS do upnp or other NAT holepunching?
<toppler>
Icefoz_: Yes.
<toppler>
But tech support is telling me that our router isn't upnp configured, and we've had it vetted by a security consultant.
<toppler>
So I want to be sure that's what is going on.
hacdias has quit [Quit: Page closed]
hacdias has joined #ipfs
<Magik6k>
toppler: IPFS does TCP hole-punching with an ICE-like protocol
<toppler>
Magik6k: I'm not sure what that means. Do you have a link?
<toppler>
Oh, wiki.
hacdias has quit [Client Quit]
<toppler>
Magik6k: I want to make sure our router is locked down against this stuff.
hacdias has joined #ipfs
<toppler>
Unless this is much weaker than actual port forwarding.
<pjz>
toppler: ICE is hard to lock down without doing full http(s) proxying, which is even *more* painful
<toppler>
pjz: Ah thanks. I'd not heard of this before.
dhruvbaldawa has joined #ipfs
domanic has quit [Ping timeout: 258 seconds]
tombusby has joined #ipfs
ylp has left #ipfs [#ipfs]
<pjz>
toppler: ICE is more or less the new name for the range of STUN/TURN protocols/methods
<Steverman>
How does the browser handle pull streams when retreiving files from another browser?
clickjack has joined #ipfs
dhruvbaldawa has quit [Remote host closed the connection]
<toppler>
pjz: Haven't heard of those either XD
<toppler>
pjz: Seems I've got some stuff to learn here.
<pjz>
toppler: ...always
<pjz>
whyrusleeping: 0.4.13 seems much better behaved wrt memory usage, I think.
<toppler>
pjz: I can't seem to access docs on eyeball.com at the moment. Any chance you can field this question: dmesg is telling me that a SYN packet addressed to my LAN IP came from an internet IP address. I assume that my router would drop such packets. So who did the NAT?
<pjz>
toppler: your router did
<pjz>
toppler: do you know how NAT works?
<toppler>
pjz: I believe so.
<toppler>
Why did my router not drop/reject the packet though?
<pjz>
toppler: how do packets that reply to NATd outbound packets get to the right host?
<toppler>
My understanding is that the router maintains information about the connection, and remaps ports and IP addresses. But this is only for outbound connections, not inbound.
<pjz>
if you look at a connection in the middle, can you tell if it's inbound or outbound?
<toppler>
pjz: Yes? Isn't that what ESTABLISHED is about in netfilter?
<pjz>
You can't - it's just packets
<toppler>
Sorry. I misread your question.
<pjz>
ESTABLISHED in netfilter means that the router has a mapping
<toppler>
This is a SYN packet, though, right?
<toppler>
My router knows the difference between an inbound SYN packet and an outbound one.
<pjz>
if it's sent to an IP/port that has a mapping, it will be forwarded, though.
<toppler>
Hmm...I didn't realise that. I assumed it would be dropped regardless.
vivus has joined #ipfs
whenisnever has quit [Quit: WeeChat 1.9.1]
dhruvbaldawa has joined #ipfs
<toppler>
So I am actually the one soliciting these connections?
mtodor has quit [Remote host closed the connection]
<pjz>
well, somehow, perhaps indirectly, but yes - unless you're being portscanned or something, of course.
dhruvbaldawa has quit [Ping timeout: 248 seconds]
Adbray has joined #ipfs
<toppler>
pjz: Do these connections then have to go over a relay?
chiui has quit [Ping timeout: 248 seconds]
James_Epp has joined #ipfs
<pjz>
toppler: I don't know what you mean
<Steverman>
I found the issue with browser <> browser, but I can't fix it! Whenever I try to .get() a file between browsers, it only works if there are no chunks!
<Steverman>
Why is that?
<Steverman>
E.g it's below the size before being chunked
<Steverman>
was it 256 kB?
upperdeck has quit [Ping timeout: 240 seconds]
<Steverman>
sorry, 2.56 mB
<Icefoz_>
Steverman: iirc the chunk size is 256k
leavehouse has joined #ipfs
<toppler>
pjz: Nevermind. I don't want to ask uninformed questions. I'll see if those docs are up again later.
<toppler>
Cheers for the help anyway.
upperdeck has joined #ipfs
<Steverman>
Aha
<Steverman>
But I still have the problem
ccii1 has joined #ipfs
ccii has quit [Ping timeout: 248 seconds]
Encrypt has joined #ipfs
upperdeck has quit [Ping timeout: 240 seconds]
f0i has joined #ipfs
upperdeck has joined #ipfs
<Steverman>
Omg, it's Chrome
disusered has joined #ipfs
totole has joined #ipfs
totole has quit [Client Quit]
totole has joined #ipfs
domanic has joined #ipfs
totole has left #ipfs [#ipfs]
tiroliro has joined #ipfs
<tiroliro>
Is there any problem with orbit chat? I mean, I can seldom recover old messages from that network.
domanic has quit [Ping timeout: 264 seconds]
<lgierth>
sidharder_: that's why `ipfs add` and `ipfs dag put` are separate commands ;) add results in a unixfs data structure with a CIDv0 (Qm...), and dag put results in a CBOR IPLD data structure with a CIDv1
<lgierth>
unixfs will migrate to "native" IPLD and CIDv1 eventually too
f0i has quit [Read error: Connection reset by peer]
domanic has joined #ipfs
upperdeck has quit [Ping timeout: 240 seconds]
upperdeck has joined #ipfs
domanic has quit [Ping timeout: 250 seconds]
domanic has joined #ipfs
tiroliro has quit [Ping timeout: 240 seconds]
ianopolous has joined #ipfs
tiroliro has joined #ipfs
upperdeck has quit [Ping timeout: 248 seconds]
upperdeck has joined #ipfs
upperdeck has quit [Ping timeout: 248 seconds]
upperdeck has joined #ipfs
rendar has quit [Quit: std::lower_bound + std::less_equal *works* with a vector without duplicates!]
domanic has quit [Ping timeout: 250 seconds]
domanic has joined #ipfs
mikedd has quit [Quit: Connection closed for inactivity]
upperdeck has quit [Ping timeout: 248 seconds]
upperdeck has joined #ipfs
<toppler>
pjz: Ack. Sorry, this is still driving me nuts. :( My local machine drops incoming packets that are not ESTABLISHED. I am seeing these being dropped in my dmesg. I still don't see how they're arriving at my machine. I thought routers generally did NAT by tracking outgoing connections and then keeping NAT mappings for the duration of those connections. If I haven't got an established connection, why am I receiving a SYN packet?
<lgierth>
or disable upnp / nat-pmp in your router
<toppler>
lgierth: If this is upnp, then that's something we need to sort out. But we've been told by tech support that our router doesn't support upnp.
<toppler>
(There's a chance tech support are wrong on this)
Winwin was banned on #ipfs by lgierth [*!*@35.100.126.78.rev.sfr.net]
Winwin was kicked from #ipfs by lgierth [Winwin]
erictapen has joined #ipfs
<lgierth>
i can only imagine the reason for those SYNs is that 1) your router has a static port forwarding set up 2) ipfs did upnp / nat-pmp 3) your router being wonky
<lgierth>
are they on port 4001 or an ephemeral port?
<toppler>
4001
<toppler>
We are definitely not supposed to be forwarding those.
<toppler>
*that
<toppler>
But yeah, sorry if this ended up being off-topic. I only noticed this with IPFS and can't explain how this is happening. If our router is misconfigured, that's a serious issue.
<lgierth>
it's a terrible single point of failure at the moment
<Steverman>
my CPU got nuked
espadrine has quit [Ping timeout: 240 seconds]
<Steverman>
and then NodeJS webrtc support is beyond bad. My solution was to add the nodejs node to the bootstrap list for browsers
<Steverman>
Not sure if that's the proper (temporary) solution
<Steverman>
I have to write a ~20-30 page report on IPFS and my project, and I am not sure how well it scales at the moment. If it actually used DHT, then I it's already well documented how well it scales
<Steverman>
on how well it scales*
<whyrusleeping>
lgierth: how hard would it be for us to spin up another bootstrapper with webrtc start?
rodolf0 has joined #ipfs
<Steverman>
Is it a fully connected mesh network for the current implementation?
<Steverman>
For browsers :)
<lgierth>
whyrusleeping: then we'd have two webrtc signallers with split brains
<whyrusleeping>
is that bad?
<lgierth>
yeeah there's just people who won't ever be able to see each other
<lgierth>
it'd work if we had working peer routing
<whyrusleeping>
:/
upperdeck has quit [Ping timeout: 248 seconds]
<whyrusleeping>
Steverman: so yeah, its a fully connected mesh right now
<lgierth>
the simplest would be if js would just be a tiny bit smarter about who to connect to x)
upperdeck has joined #ipfs
<Steverman>
Gotcha
<lgierth>
and content routing will be awesome to have too -- right now js-ipfs tightly couples bitswap to the bootstrappers
<lgierth>
i.e. the js default bootstrappers == the gateways
<Steverman>
Whelp, I can already foresee how my statistical analysis on the scalability will be like
<lgierth>
Steverman: there's so much more potential
<lgierth>
we know what the problems are, and how to fix them
<lgierth>
but a day has only 24h
<Steverman>
:D
<Steverman>
I know. It's my choice for using js-ipfs in its infancy
<Steverman>
I could drop browsers all together, but where's the fun in that if people can't join in on a live demo
upperdeck has quit [Ping timeout: 268 seconds]
<Steverman>
But I haven't actually tested 50~ peers in my network
upperdeck has joined #ipfs
<Steverman>
that's how many students that'll be joining in my live demo for my p2p course
<whyrusleeping>
we had around 30 people join in for a demo of peerpad
<whyrusleeping>
and it went well enough
<Steverman>
But I am sending video segments for live streaming
<whyrusleeping>
hrm
<timuckun>
Is it possible to set up a "private" ipfs system where data is only shared amongst an organization?
<whyrusleeping>
timuckun: yeap, look at ipfs private networks
<timuckun>
@whyrusleeping I see some github issues. Is this implemented yet?
<Steverman>
How about data requests in browsers. Is it still "Everyone! Send me this file if you have it"?